Scraper.is MCP
Scraper.is MCP enables AI assistants to scrape web content in multiple formats with real-time updates. It integrates seamlessly via the MCP protocol for up-to-date web data access.
How to Install and Use Scraper.is MCP
Scraper.is MCP is a handy tool that helps AI assistants scrape web content using the Model Context Protocol (MCP). It lets you get information from the web in different formats like markdown, HTML, JSON, or even screenshots. If you want to try this tool, here’s a simple guide on how to install it and start using it with easy code examples.
Step 1: Prepare Your Environment
Before installing Scraper.is MCP, you need an API key from Scraper.is. This key lets the tool access the web scraping service securely.
To get your API key:
- Go to scraper.is and sign up or log in.
- Open your dashboard and visit the API Keys section at https://www.scraper.is/dashboard/apikeys.
- Create a new API key or copy an existing one.
- Keep this API key safe because you will need it soon.
After you have your key, create a .env file in your project folder and add your key like this:
SCRAPERIS_API_KEY=your_api_key_here
Replace your_api_key_here with your real key.
Step 2: Install Scraper.is MCP
Next, you need to install the Scraper.is MCP package on your computer. You can use either npm or yarn for this.
If you use npm, open your terminal or command prompt and run:
npm install -g scraperis-mcp
If you prefer yarn, use this command:
yarn global add scraperis-mcp
The -g or global option installs the tool so you can run it from anywhere on your system.
Step 3: Set Up for Claude Desktop (Optional)
If you want to use Scraper.is MCP with Claude Desktop, you have to add a small configuration.
After installing, locate your claude_desktop_config.json file. Add the following inside the "mcpServers" section:
{
"mcpServers": {
"scraperis_scraper": {
"command": "scraperis-mcp",
"args": [],
"env": {
"SCRAPERIS_API_KEY": "your-api-key-here",
"DEBUG": "*"
}
}
}
}
Change "your-api-key-here" to your actual API key. Then restart Claude Desktop to apply the changes.
Step 4: Use Scraper.is MCP with AI Assistants
The main feature of Scraper.is MCP is the scrape tool. It lets AI assistants grab content from websites based on your prompt and the output format you want.
Here is the basic JSON format when calling the scrape tool:
{
"prompt": "Get me the top 10 products from producthunt.com",
"format": "markdown"
}
You can choose several formats for the answer:
markdown: The content will be in markdown style.html: You'll get the HTML code of the page.screenshot: A picture of the web page is returned.json: The content is structured in JSON.quick: A short summary.
This simple format makes it easy to tell the AI what to scrape.
Step 5: Run and Test with MCP Inspector
If you want to test or develop with Scraper.is MCP, you can use the MCP Inspector tool. It helps you see how the MCP servers communicate.
Run this command in your terminal:
npx @modelcontextprotocol/inspector scraperis-mcp
This will start the inspector using the Scraper.is MCP server for testing.
Now you know how to install Scraper.is MCP, set it up with your API key, and use it with AI assistants or Claude Desktop. With this, you can quickly get web content in many formats right from your tools. Happy scraping!